Enhancement of photocatalytic activity for WO3 by simple NaOH loading

Article abstract of DOI:10.1016/j.apcata.2014.09.046

We found a simple method for preparation of NaOH-loaded WO₃ with high photocatalytic activity under visible light irradiation. The sample was characterized with XRD, UV–Vis, VB-XPS, and other methods. From the VB-XPS data, the potentials of bottom of conduction band (CB) and top of valence band (VB) for WO₃ were found to be lifted up by NaOH loading. Furthermore, this NaOH-loaded WO₃ exhibited marked photocatalytic activity for 2-propanol decomposition into CO₂ under visible light irradiation. The activity of 5 wt% of NaOH-loaded WO₃ was 120 times higher than that of pure WO₃. Possible reasons for the higher activity of the NaOH-loaded WO₃ are the raising of the potential of CB and suppression of H₂O₂ accumulation. More detailed analyses were conducted using the reaction mechanism and potentials of VB and CB. This stable solid-base-loaded WO₃ maintains its photocatalytic activity for over 3 months.

Full text of DOI:10.1016/j.apcata.2014.09.046

Applied Catalysis A: General 488 (2014) 183–188
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Enhancement of photocatalytic activity for WO3 by simple NaOH
loading
Tetsuya Kakoa,∗, Xianguang Menga,b, Jinhua Yea,b
a Environmental Remediation Materials Unit, National Institute for Materials Science (NIMS), 1-1 Namiki, Tsukuba, Ibaraki 305-0044, Japan
b Graduate School of Chemical Science and Engineering, Hokkaido University, Sapporo 060-0814, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
We found a simple method for preparation of NaOH-loaded WO3 with high photocatalytic activity under
visible light irradiation. The sample was characterized with XRD, UV–Vis, VB-XPS, and other methods.
From the VB-XPS data, the potentials of bottom of conduction band (CB) and top of valence band (VB) for
WO3 were found to be lifted up by NaOH loading. Furthermore, this NaOH-loaded WO3 exhibited marked
photocatalytic activity for 2-propanol decomposition into CO2 under visible light irradiation. The activity
of 5 wt% of NaOH-loaded WO3 was 120 times higher than that of pure WO3. Possible reasons for the
higher activity of the NaOH-loaded WO3 are the raising of the potential of CB and suppression of H2O2
accumulation. More detailed analyses were conducted using the reaction mechanism and potentials of
VB and CB. This stable solid-base-loaded WO3 maintains its photocatalytic activity for over 3 months.
© 2014 Elsevier B.V. All rights reserved.
Received 30 June 2014
Received in revised form
25 September 2014
Accepted 29 September 2014
Available online 7 October 2014
Keywords:
Solid base
Band structure
H2O2
Bottom of conduction band
Electron consumption.
1. Introduction
visible-light irradiation [17,18]. However, its photocatalytic activ-
ity for gaseous organic decomposition is limited [19]. Therefore,
Large amounts of harmful organic compounds of many kinds
exist in our daily life, and these harmful ones often damage
our health. Such harmful organic compounds must be removed
from the environment to protect our safe and comfortable life.
Photocatalysis is a useful technology for removing them [1,2]. In
photocatalysis, harmful organics are decomposed mainly by pho-
togenerated holes. However, if photogenerated electrons are not
consumed, these electrons will recombine easily with holes, lead-
ing to a decrease in the photocatalytic activity or deactivation of a
photocatalyst.
investigations of several kinds have been undertaken to enhance
its photocatalytic activity [20–23]. Indeed, co-catalyst loading
including Pt or Pd is a good method for higher activity [24,25].
However, because Pt and Pd are precious metals, these metals can-
not be used in the WO3 photocatalyst in large amounts. These
precious metals probably deactivates easily by catalyst poison-
ing of sulphur-containing gases such as H2S and CH3SH in human
living environments [26]. Consequently, in terms of costs, visible-
light-sensitive photocatalysts of different types are also needed for
varieties of the photocatalysts.
Photocatalysis functioning under visible light irradiation is
regarded as a promising technology for the removal of indoor harm-
ful organics because indoor illumination contains large amounts of
visible light. At present, the activity of the reported photocatalysts
is limited under visible-light irradiation, and novel visible-light-
sensitive photocatalysts with higher activity have been expected.
Therefore, development of novel photocatalysts has been studied
intensively [3–16].
The addition of alkali hydroxide enhances the photocatalytic
water-splitting property over Pt-loaded TiO2- and SrTiO3-related
materials [27–30]. This enhancement reportedly suppresses
reverse reactions (2H2 + O2 = 2H2O) on the Pt surface by alkali
addition [27,31]. However, this enhancement mechanism is not
always applicable for the case of photocatalytic organic decompo-
sition because the reaction system for water splitting differs greatly
from that for organic decomposition. Furthermore, photocatalytic
decomposition of gaseous organics in alkali condition has been
reported slightly. It should be intensively studied.
As a binary-oxide semiconductor, WO3 shows high activity for
photocatalytic O2 evolution from an aqueous AgNO3 solution under
Very recently, Aminian and Hakimi partially reported some
effects of AOH (A = Na, K, and Cs) on photocatalytic activity for
WO3 [32]. According to his report, 0.075 wt% of KOH-loaded WO3
showed about four times higher activity than pure WO3 for gaseous
∗
Corresponding author. Tel.: +81 29 859 2848; fax: +81 29 860 4958.
E-mail address: kako.tetsuya@nims.go.jp (T. Kako).
http://dx.doi.org/10.1016/j.apcata.2014.09.046
0926-860X/© 2014 Elsevier B.V. All rights reserved.

184
T. Kako et al. / Applied Catalysis A: General 488 (2014) 183–188
organic decomposition., However, they also described that too
much AOH loading, such as 5 wt% of loading, decreased the activ-
ity of WO3 largely. In addition, Wada et al. described that WO3
resolves well in the solution with high NaOH concentration, includ-
ing 1 mol/L of an aqueous NaOH solution, for a short time. Its
application is extremely limited because of this dissolution [33].
So, the effect of a large amount of NaOH loading, such as over
5 wt% of loading, on the WO3 photocatalytic activity and stabil-
ity remains unclear. More experiments are needed. Therefore, we
examined photocatalytic activity of 5 wt% of NaOH-loaded WO3
under visible-light irradiation in this study. Especially, we carefully
prepared 5 wt% of NaOH-loaded WO3 by controlling the prepara-
tion condition including heating temperature. Results show that
our prepared NaOH (solid base)-loaded WO3 showed marked pho-
tocatalytic activity and high durability for organics decomposition
in the gas phase.
measured using a gas chromatograph (GC-14B; Shimadzu) with
flame-ionized detectors (FID) and a methanizer.
The specific surface area was evaluated with a surface area ana-
lyzer (Gemini-2460; Micromeritics Instrument Corp., USA) using
Brunauer-Emmett-Teller (BET) method at 77 K. Furthermore, the
areas were 5.8, 7.0, 8.6, and 4.5 m2/g, respectively, for pure WO3,
5 wt%-NaOH-WO3, 25 wt%-NaOH-WO3, and 0.075 wt%-KOH-WO3.
3. Results and discussion
3.1. XRD
Fig. 1 presents the XRD patterns of pure WO3 and 5 wt% of
NaOH-loaded WO3. From the XRD results and data from an ear-
lier report [34], the 5 wt% of NaOH-loaded WO3 mainly consists of
monoclinic WO3 (PDF No. 43-1035). A small amount of crystallized
Na2WO4·2H2O with orthorhombic structure (PDF No. 13-0431)
also exists in the sample. Reportedly, WO3 can partially react with
NaOH. Thereby, Na2WO4·2H2O can be formed [35].
2. Experimental
2.1. Material preparation
3.2. Optical absorption property
Powder of NaOH-loaded WO3 was prepared by the following
method: 10 mL of an aqueous NaOH (Wako Pure Chemical Indus-
tries Ltd., Japan) solution (NaOH concentration: 20 g/L) and 4 g of
monoclinic WO3 (Wako) [34] were mixed on a mortar for over
30 min, and the mixed powder was dried in an oven at 343 K for
4–5 h. Finally, we obtained the sample, of which the mixing ratio
of NaOH to WO3 was 5 wt% (NaOH: WO3 = 5:100). Additionally,
25 wt% of NaOH-loaded WO3 was prepared by using this method.
As a reference sample, 0.075 wt% of KOH-loaded WO3 powder was
prepared using the method described in the earlier report. This
powder was obtained by heating of the mixed KOH-WO3 powder
at 723 K for 2 h [32].
As described above, the new phase was formed on the sample
(NaOH-loaded WO3) and this formation might affect the absorp-
tion property. Therefore, optical absorption spectra were measured
using a UV–vis spectrophotometer. We checked an effect of the
NaOH loading on the optical absorption property. Fig. 2 presents the
optical absorption spectra of WO3 and the 5 wt% of NaOH-loaded
WO3. Both samples can absorb visible light well. Furthermore,
2.2. Material characterization
The prepared samples were first characterized with an X-ray
diffraction meter (XRD, X’pert Pro; PANalytical Co., Netherlands)
with Cu K␣ radiation. Using the Kubelka–Munk relationship, optical
absorption spectra of the prepared samples were converted from
their respective reflectance spectra, which were measured using a
UV–Vis spectrophotometer (UV-2500PC; Shimadzu Corp. Japan).
Results of valence-band X-ray photoemission spectroscopic
data (VB-XPS) were obtained by measuring the samples using
an X-ray photoelectron spectrometer (AXIS-HS; Shimadzu-Kratos
Analytical Com., Japan) with an X-ray source of monochromatic Al.
Binding energy was calibrated using C 1s peak with binding energy
of 284.5 eV.
Fig. 1. XRD patterns of pure WO3 and 5 wt% of NaOH-loaded WO3. Arrows indicate
the XRD patterns of Na2WO4·2H2O.
Photocatalytic activity at room temperature was evaluated from
2-propanol (IPA) into CO2 via acetone. Details of the evaluation
method are the following: the powder sample with weight of 0.4 g
spread evenly on the dish with the area of about 8 cm2. The dish
was set on the glass reactor with a volume of 500 cm3. The inside
atmosphere of the reactor was replaced with pure air. Further-
more, IPA gas was injected into the reactor. Then, the concentration
of the IPA in the reactor was estimated as 700–800 ppm. Subse-
quently, the reactor was kept in the dark until the IPA gas reached
the adsorption–desorption equilibrium state. Then, the reactor was
irradiated with visible light (400 nm < ꢀ < 530 nm), which was emit-
ted from a 300-W Xe lamp equipped with a water filter and glass
filters of three kinds [UV-cutoff filter (Y-44), blue filter (B390), and
IR cutoff filter (HA-30), Hoya. Corp., Japan]. The light intensity was
about 1 mW cm−2, which was evaluated using a spectroradiometer
(UV-40, Ushio Inc., Japan). The IPA, acetone, and CO2 gases were
Fig. 2. Optical absorption spectra of WO3 and 5 wt% of NaOH-loaded WO3.

T. Kako et al. / Applied Catalysis A: General 488 (2014) 183–188
185
Fig. 4. Schematic illustration of proposed band structure of WO3 and 5 wt% of NaOH-
loaded WO3.
Fig. 3. VB-XPS spectra of WO3 and 5 wt% of NaOH-loaded WO3.
results show that the absorption spectrum of the NaOH-loaded
WO3 slightly blue-shifted compared with that of WO3. This blue
shift is possibly attributable to the change in surface functional
groups [36,37] and existence of other compounds including NaOH
with white colour.
Next, the absorption properties were evaluated qualitatively to
elucidate their differences in the properties. First, the band gaps
were calculated from the onset absorption edges [38]. The band
gaps of pure WO3 and the NaOH-loaded WO3 were estimated,
respectively, as 2.6 and 2.7 eV. As a reference, the absorption spec-
trum of Na2WO4·2H2O was also measured and the band gap was
about 4.1 eV. Its band gap is over 1 eV larger than that of the NaOH-
WO3, so formation of Na2WO4 does not affect the VB-XPS data
(described in Section 3.3) around the top of VB for the NaOH-WO3.
Fig. 5. Photocatalytic IPA decomposition into acetone at the initial stage under
visible-light irradiation over WO3 and 5 wt% of NaOH-loaded WO3.
3.3. Band structure
The band structure of a photocatalyst usually affects its pho-
tocatalytic activity. Therefore, information related to the band
structure is important. Consequently, the potentials for top of the
valence band (VB) for the samples were estimated. Then, the band
structures were discussed using these estimated potentials and the
band gaps.
would be at least 0.4 V lower than that of pure WO3. This value is
close to that of the shift (0.5 eV), which was measured by VB-XPS.
Therefore, we consider that this kind of the shift may occur on the
base (NaOH)-coated photocatalyst surface.
3.4. Photocatalytic activity
First, the potentials for top of the VB were estimated using VB-
XPS technique. Fig. 3 presents the XPS spectra of the samples at
around 0 eV in the binding energy. This result suggests that the
potential for the top of VB for the NaOH-loaded WO3 was 0.5 eV
smaller than that for WO3. Furthermore, the potential for top of VB
for WO3 can be estimated as +3.0 V (vs SHE) because the poten-
tial for bottom of CB was reported as +0.4 V (vs SHE) [39] and its
band gap was 2.6 eV. Therefore, the potentials for the top of VB and
bottom of CB for the NaOH-loaded WO3 were estimated as –0.2
and +2.5 V (vs SHE), respectively. Based on these results, the band
structure can be written as shown in Fig. 4.
Photocatalytic activity was evaluated from IPA decomposition
into CO2 via acetone because IPA is a frequently used VOC in indus-
try. In addition, the IPA decomposition is a standard reaction for
photocatalytic-activity evaluation [11,15]. First, we evaluated IPA
decomposition into acetone by photogenerated holes at the initial
reaction stage. Fig. 5 shows a change in concentrations of acetone
generation in IPA decomposition between 0 and 2 h of visible-light
irradiation. Between 0 and 40 min, the acetone concentration over
WO3 is almost equal to that over the NaOH-loaded WO3. Subse-
quently, however, pure WO3 was partially deactivated. The acetone
concentration for WO3 is half that for the NaOH-loaded WO3 at
2 h of light irradiation. This result suggests that the NaOH loading
does not work as enhancement of WO3 photocatalytic activity itself.
However, the loading functions as suppression of deactivation
for WO3 activity. By further visible light irradiation, intermediate
acetone can be decomposed into the final product CO2 by photogen-
erated holes. Results show more marked difference in the activity
between WO3 and the NaOH-loaded WO3, as shown in Fig. 6. The
figure shows the time dependence of changes of acetone and CO2
concentrations over WO3 and the NaOH (5 wt%)-loaded WO3. For
the NaOH-loaded WO3, acetone was generated mainly. The con-
centration of acetone generation increased linearly with irradiation
time of 0–24 h. After about 24 h, the concentration of acetone in the
The reason why the potential for top of VB is less positive by
NaOH loading remains unclear, but a possible reason is the follow-
ing. The potential of top of VB has been reported as less positive in
an alkali solution. This shift obeys the relation presented in Eq. (1).
V0 = V0(pH 0) − 0.06pH
(1)
Here V0 denotes the potential of the top of VB [40]. This relation
might be extended to solid-base-loaded semiconductor (photocat-
alyst). For 5 wt%-NaOH-WO3, the surface is basic. When the sample
was prepared, an aqueous NaOH solution (pH >13) was used. For
pure WO3, the surface is neutral. It is surrounded with air and
adsorbed water (pH ≈ 7). If the shift above (Eq. (1)) occurs in this
system, then the potential of the top of VB of 5 wt%-NaOH-WO3

186
T. Kako et al. / Applied Catalysis A: General 488 (2014) 183–188
Fig. 6. Photocatalytic IPA decomposition into CO2 via acetone between 0 and 200 h
under visible-light irradiation over WO3 and 5 wt% of NaOH-loaded WO3.
Fig. 7. Repeated tests for photocatalytic decomposition of IPA into CO2 over 25 wt%
of NaOH-loaded WO3 under visible-light irradiation. 600–800 ppm of IPA gas in
every turn except for the second run was injected into the reactor. In the second
run, 860 ppm of IPA was injected.
gas phase decreased and CO2 concentration increased drastically
and linearly, indicating that intermediate acetone was decomposed
into CO2. Furthermore, after about 200 h of visible-light irradiation,
the injected IPA was decomposed almost completely into CO2 by
photogenerated holes for the NaOH-loaded WO3. For WO3, the con-
centrations of acetone and CO2 generation increased slowly with
irradiation time. The concentrations were extremely low. When the
photocatalytic activity of the NaOH-loaded WO3 is compared with
that of pure WO3 using CO2 concentrations after 160 h, the activity
of the NaOH (5 wt%)-loaded WO3 was over 120 times higher than
that of pure WO3. Larger surface area contributed little to higher
photocatalytic activity of 5 wt%-NaOH-WO3 because the surface
area of the 5 wt%-NaOH-WO3 (7.0 m2/g) was only 1.2 times higher
than that of pure WO3 (5.8 m2/g). The reason for this higher surface
area is expected to derive from NaOH coverage of the 5 wt%-NaOH-
WO3 surface.
The NaOH-loaded WO3 shows much higher activity than WO3,
which might be attributable to the ease of consumption for pho-
togenerated electrons. Apparently, the reactions for consumption
of photogenerated electrons on the NaOH-loaded WO3 should be
different from those for WO3. For WO3, photogenerated electrons
can be reacted with O2 through two-electron oxygen reduction (Eq.
(4)), and H2O2 is formed [41]. However, H2O2 is saturated quickly
and the rate for H2O2 consumption is very slow (ESI, Fig. S1) [41],
indicating that H2O2 formation almost stops on the WO3 after satu-
ration. Because the other electron consumption cannot occur, little
electron consumption occurs at H2O2 saturation. Therefore, the
photogenerated electrons are accumulated on the WO3. The elec-
trons recombine easily with holes, thereby greatly decreasing the
activity quickly.
H2O2 becomes more reactive [43]. It is rapidly decomposed into
reactive OOH− (Eq. (5)), which further reacts with the photoexcited
electrons (Eq. (6)). Therefore, for NaOH-loaded WO3, H2O2 can be
consumed more easily via Eqs. (5)–(7) [44]. The rate of consump-
tion of H2O2 for NaOH-loaded WO3 is expected to be much higher
than that for pure WO3, leading to higher activity of NaOH-loaded
WO3. This expectation is supported by measurement of the H2O2
amount on the photocatalyst surface; the amount of H2O2 on the
NaOH-WO3 was much smaller than that on pure WO3 (ESI, Fig. S3).
O2 + H+ + e− = ·OOH (−0.046 V vs SHE)
O2 + e− = ·O2− (−0.28 V vs SHE)
O2 + 2H+ + 2e− = H2O2 (+0.68 V vs SHE)
H2O2 = H+ + OOH−
(2)
(3)
(4)
(5)
(6)
(7)
(8)
OOH− + H2O + 2e− = 3OH− (+0.87 V vs SHE)
2OOH− = O2 + 2OH−
H
+ + OH− = H2O
As described above, the 5 wt% of NaOH-loaded WO3 showed rel-
atively high activity. Furthermore, a larger amount of NaOH-loaded
sample, 25 wt% of NaOH-loaded WO3 showed visible-light sensitiv-
ity (ESI, Fig. S2) and relatively high activity. Its activity resembles
the activity of 5 wt% of NaOH-loaded WO3. In addition, 25 wt% of
NaOH-loaded WO3 consisting of WO3, Na2WO4, and NaOH (ESI,
Figs. S3 and S4) shows stable photocatalytic performance for a long
time, as shown in Fig. 7. The figure shows repeated tests for CO2
generation in photocatalytic IPA decomposition under visible-light
irradiation over 25 wt% of NaOH-loaded WO3. Based on this result,
we found that the photocatalyst maintains its relatively high activ-
ity for about 3 months. Furthermore, if 1 mol of IPA (C3H8O) is
considered to be decomposed into 3 mol of CO2 by 18 holes [45,46],
just as Eq. (9), the turnover number (TON) of this photocatalyst
became about 1.1 after the eighth run of the photocatalytic repeated
tests. Therefore, we conclude that the IPA decomposition over the
NaOH-loaded WO3 should be regarded as a genuine photocatalytic
reaction.
For the NaOH-loaded WO3, the electrons can be consumed by
one-electron or two-electron O2 reductions. These reduction reac-
tions engender ease of consumption for photogenerated electrons,
as described below. As shown in Fig. 4, the potential for the bottom
of CB for the NaOH-loaded WO3 is more negative than potentials
for one-electron O2 reduction (Eqs. (2) and (3)) [24]. Therefore,
one-electron O2 reduction may occur. The rate of this reduction
is regarded as much higher than that of the two-electron oxygen
reduction (Eq. (4)) or H2O2 consumption reaction [42], leading to
higher activity of the NaOH-loaded WO3. To elucidate this proposed
mechanism, we will plan to systematically investigate the forma-
C3H8O + 5H2O + 18h+ = 3CO2 + 18H+
(9)
−
tion of ·O2 or ·OOH radical by ESR or other technique. We will
discuss it in a future report.
The results of the activity and heating temperature reported
herein completely differ from those of previously reported study
by Aminian et al. [32]. In that study, 5 wt% of alkali loading at 723 K
was too large. In addition, loading of too much alkali hydroxide
cannot enhance the photocatalytic activity of WO3 because of the
complete phase transition and morphology changes from nanopar-
ticles into nanorods. Furthermore, 0.075 wt% of KOH-loaded WO3
Aside from one-electron O2 reduction, two-electron O2 reduc-
tion (Eqs. (4) and (6)) is expected to occur on the NaOH-loaded WO3.
For the reduction described in Eq. (4), one must consider whether
H2O2 is saturated on the surface of the NaOH-loaded WO3. This is
because H2O2 saturation decreases the photocatalytic activity eas-
ily and heavily, just as WO3. However, under alkaline environment,

T. Kako et al. / Applied Catalysis A: General 488 (2014) 183–188
187
2-propanol decomposition and stable photocatalytic performance
of the NaOH-loaded WO3.
IPA is mainly decomposed by photogenerated holes. However,
if photogenerated electrons do not consume well, then recom-
bination between holes and electrons occurs easily, leading to
deactivation or low activity of the photocatalyst. Therefore, con-
sumption of electrons is extremely important. For pure WO3,
electrons are first consumed through two-electron O2 reduction
and H2O2 forms. However, H2O2 cannot be decomposed well and
saturated on the surface, leading to decrease in the activity. How-
ever, for NaOH-loaded WO3, electrons can be consumed by two
reactions including one-electron and two-electron O2 reduction.
For two-electron O2 reduction, H2O2 may form, but this H2O2 is
decomposed more easily because H2O2 becomes reactive in alkali
conditions. For one-electron O2 reduction, H2O2 does not form and
accumulate on the surface, leading to maintenance of its activity.
Additionally, results show that the activity of NaOH (solid
base)-loaded WO3 is affected strongly by heating temperature and
loading amount of NaOH. Low heating temperature and a large
loading amount engender higher photocatalytic activity. Results
show that the 5 wt% of NaOH-loaded WO3 heated at 343 K for this
study showed over four times higher activity than the 0.075 wt% of
KOH-loaded WO3 described in the earlier report. In addition, 25 wt%
of NaOH-loaded WO3 shows relatively high photocatalytic activity
and retains its photocatalytic activity for about 3 months. Based
on these results, we conclude that solid-base loading is a promis-
ing method for the development of stable photocatalysts with high
activity.
Fig. 8. Photocatalytic IPA decomposition into CO2 under visible-light irradiation
over WO3, 5 wt% of NaOH-loaded WO3, and 0.075 wt% of KOH-loaded WO3.
prepared at 723 K reportedly showed the highest activity among
the KOH-loaded WO3 samples, with loading amounts of 0–5 wt%
reported for their study. Therefore, we prepared 0.075 wt% of KOH-
loaded WO3 as a reference sample and compared its photocatalytic
activity with the activity of our 5 wt% of NaOH-loaded WO3 to eluci-
date the activity difference between this study and the previously
reported studies. Fig. 8 shows the time dependence of change of
CO2 generation in IPA decomposition under visible-light irradia-
tion in the presence of some photocatalysts. Indeed, IPA can be
decomposed by photocatalysis of 0.075 wt% of KOH-loaded WO3.
However, the activity for the 0.075 wt% of KOH-loaded WO3 was
over four times lower than that for the 5 wt% of NaOH-loaded WO3
when the activity was evaluated from concentrations of CO2 gen-
eration after about 160 h of light irradiation. The higher activity of
the 5 wt%-NaOH-loaded WO3 was first considered. Since the mor-
phology changes were reported to lead to decrease in the activity
as described above [32], the morphology of pure WO3 and these
two alkali-treated WO3 was measured using SEM, which revealed
no marked difference (ESI, Fig. S4). These three samples were
composed entirely of nanoparticles, indicating that morphology
changes that led to decreased activity did not occur by 5 wt%-NaOH-
loading at 343 K. Namely, results show that 5 wt%-NaOH-loading
at low temperature did not change the morphology. In addition,
because a larger amount of alkali loading can promote electron
consumption as described above, the 5 wt%-NaOH-loaded WO3 can
show higher activity than the 0.075 wt%-KOH-loaded WO3. Aside
from these factors, the larger surface area contributed slightly to
higher activity of the 5 wt%-NaOH-loaded WO3. The activity for
the 0.075 wt% of KOH-loaded WO3 annealed at 723 K was report-
edly much higher than that for 5 wt% of KOH-loaded WO3 at 723 K.
Therefore, it can be inferred that 5 wt% of NaOH-loaded WO3 heated
at 343 K shows much higher activity than 5 wt% of KOH-loaded
WO3 at 723 K with phase-transferred nanorods. From these results,
we infer that the heating temperature is also an important factor
for higher activity. Furthermore, we concluded that an amount of
NaOH loaded should be large (5–25 wt%). The loaded sample should
be dried at low temperature, such as 343 K and room temperature,
for development of NaOH-loaded WO3 with higher activity.
Appendix A. Supplementary data
Supplementary data associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/j.apcata.
2014.09.046.
References
[1] M.R. Hoffmann, S.T. Martin, W.Y. Choi, D.W. Bahnemann, Chem. Rev. 95 (1995)
69–96.
[2] K. Hashimoto, H. Irie, A. Fujishima, Jpn. J. Appl. Phys. 44 (2005) 8269–8285.
[3] M. Liu, X. Qiu, M. Miyauchi, K. Hashimoto, J. Am. Chem. Soc. 135 (2013)
10064–10072.
[4] T. Kako, J. Ye, J. Mol. Catal. A 320 (2010) 79–84.
[5] H.G. Kim, D.W. Hwang, J.S. Lee, J. Am. Chem. Soc. 126 (2004) 8912–8913.
[6] A. Tanaka, K. Hashimoto, H. Kominami, J. Am. Chem. Soc. 134 (2012)
14526–14533.
[7] N. Murakami, T. Matsuo, T. Ohno, Catal. Commun. 12 (2011) 341–344.
[8] H. Irie, S. Miura, K. Kamiya, K. Hashimoto, Chem. Phys. Lett. 457 (2008) 202–205.
[9] J.Z. Bloh, R. Dillert, D.W. Bahnemann, Environ. Sci. Pollution Res. 19 (2012)
3688–3695.
[10] Z.G. Yi, J.H. Ye, N. Kikugawa, T. Kako, S.X. Ouyang, H. Stuart-Williams, H. Yang,
J.Y. Cao, W.J. Luo, Z.S. Li, Nat. Mater. 9 (2010) 559–564.
[11] R. Asahi, T. Morikawa, T. Ohwaki, K. Aoki, Y. Taga, Science 293 (2001) 269–271.
[12] T. Kako, N. Umezawa, K. Xie, J. Ye, J. Mater. Sci. 48 (2013) 108–114.
[13] S.U.M. Khan, M. Al-Shahry, W.B. Ingler, Science 297 (2002) 2243–2245.
[14] D.F. Wang, T. Kako, J. Ye, J. Am. Chem. Soc. 130 (2008) 2724–2725.
[15] T. Kako, Z.G. Zou, M. Katagiri, J.H. Ye, Chem. Mater. 19 (2007) 198–202.
[16] W. Macyk, G. Burgeth, H. Kisch, Photochem. Photobiol. Sci. 2 (2003) 322–328.
[17] T. Kako, N. Kikugawa, J.H. Ye, Catal. Today 131 (2008) 197–202.
[18] A. Kudo, K. Ueda, H. Kato, I. Mikami, Catal. Lett. 53 (1998) 229–230.
[19] T. Arai, M. Yanagida, Y. Konishi, Y. Iwasaki, H. Sugihara, K. Sayama, Catal. Com-
mun. 9 (2008) 1254–1258.
[20] K. Katsumata, R. Motoyoshi, N. Matsushita, K. Okada, J. Hazardous Mater. 260
(2013) 475–482.
4. Conclusion
[21] L. Huang, H. Xu, Y. Li, H. Li, X. Cheng, J. Xia, Y. Xu, G. Cai, Dalton Trans. 42 (2013)
8606–8616.
[22] A.J.E. Rettie, K.C. Klavetter, J.F. Lin, A. Dolocan, H. Celio, A. Ishiewkene,
H.L. Bolton, K.N. Pearson, N.T. Hahn, C.B. Mullins, Chem. Mater. 26 (2014)
1670–1677.
[23] C.W. Dunnill, S. Noimark, I.P. Parkin, Thin Solid Films 520 (2012) 5516–5520.
[24] R. Abe, H. Takami, N. Murakami, B. Ohtani, J. Am. Chem. Soc. 130 (2008)
7780–7781.
[25] T. Arai, M. Horiguchi, M. Yanagida, T. Gunji, H. Sugihara, K. Sayama, Chem.
Commun. (2008) 5565–5567.
We developed 5 wt% of NaOH-loaded WO3 sample, which was
prepared by drying the mixture of NaOH and WO3 at 343 K.
This sample showed marked photocatalytic activity for 2-propanol
decomposition into CO2 under visible light irradiation. The activ-
ity of this sample was 120 times higher than that of pure WO3.
This higher activity might derive from ease of photogenerated elec-
trons for the NaOH-loaded WO3, judging from the initial stage of

188
T. Kako et al. / Applied Catalysis A: General 488 (2014) 183–188
[26] T. Kako, H. Irie, K. Hashimoto, J. Photochem. Photobiol. A 171 (2005) 131–135.
[27] S. Sato, J.M. White, J. Catal. 69 (1981) 128–139.
[28] S.X. Ouyang, H. Tong, N. Umezawa, J. Cao, P. Li, Y. Bi, Y. Zhang, J. Ye, J. Am. Chem.
Soc. 134 (2012) 1974–1977.
[36] G. Li, T. Kako, J. Ye, J. Phys. Chem. Solid 69 (2008) 2487–2491.
[37] Z. Li, T. Yu, Z. Zou, J. Ye, Appl. Phys. Lett. 88 (2006) 071917.
[38] A. Ishikawa, T. Takata, T. Matsumura, J.N. Kondo, M. Hara, H. Kobayashi, K.
Domen, J. Phys. Chem. B 108 (2004) 2637–2642.
[29] K. Domen, S. Naito, T. Onishi, K. Tamaru, Chem. Phys. Lett. 92 (1982) 433–434.
[30] F.T. Wagner, G.A. Somorjai, Nature 285 (1980) 559–560.
[31] M. Kitano, M. Hara, J. Mater. Chem. 20 (2010) 627–641.
[32] M.K. Aminian, M. Hakimi, Catal. Sci. Technol. 4 (2014) 657–664.
[33] M. Wada, N. Wang, Y. Konishi, Y. Miseki, T. Gunji, K. Sayama, Chem. Lett. 42
(2013) 395–397.
[39] S.J. Hong, S. Lee, J.S. Jang, J.S. Lee, Energy Environ. Sci. 4 (2011) 1781–1787.
[40] M.A. Alpuche-Aviles, Y. Wu, J. Am. Chem. Soc. 131 (2009) 3216–3224.
[41] J. Kim, C. Lee, W. Choi, Environ. Sci. Technol. 44 (2010) 6849–6854.
[42] O. Tomita, R. Abe, B. Ohtani, Chem. Lett. 40 (2011) 1405–1407.
[43] A.G. Leigh, US Patent 4,304,762A (1981).
[44] E. Yeager, J. Mol. Catal. 38 (1986) 5–25.
[34] C. Zhao, Y. Yang, Z. Zhang, Open. J. Appl. Sci. 2 (2012) 86–92.
[35] Z. Mirski, T. Piwowarczyk, Arch. Civ. Mech. Eng. 11 (2011) 411–419.
[45] H. Irie, Y. Watanabe, K. Hashimoto, J. Phys. Chem. B 107 (2003) 5483–5486.
[46] H. Shi, G.Q. Chen, Z.G. Zou, Appl. Catal. B 156–157 (2014) 378–384.